Update README.md

README.md

@@ -4,4 +4,37 @@ datasets:
 - zwhe99/DeepMath-103K
 language:
 - en
----
+---
+
+## Quick Start
+
+This repository contains remote code and weights for a **Native Sparse Attention** distillation of [DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/deepseek-ai/DeepSeek-RL-Distill-Qwen-1.5B), distilled on mathematical reasoning data.
+
+### Installation
+
+To use this model, please ensure the following dependencies are installed:
+
+#### 1. Install the required sparse attention library from our custom fork:
+```bash
+pip install git+https://github.com/fnite1604/native-sparse-attention-pytorch.git
+```
+
+#### 2. Install other standard dependencies:
+These are handled automatically by the Transformers library and include:
+```bash
+pip install transformers torch
+```
+
+Note: We recommend using Python 3.8+ and PyTorch 2.0+ for compatibility.
+
+### Example Usage
+
+A `quick_start.py` script is included to help you get started with inference:
+
+```bash
+python quick_start.py
+```
+
+This will load the model and generate text based on a predefined prompt using Native Sparse Attention.
+
+